checkpoint-2390/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 2390,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.0921465968586386,
      "grad_norm": 77.68479919433594,
      "learning_rate": 7.91213389121339e-06,
      "loss": -86.9118,
      "step": 500
    },
    {
      "epoch": 4.184293193717277,
      "grad_norm": 135.78587341308594,
      "learning_rate": 5.820083682008368e-06,
      "loss": -151.7338,
      "step": 1000
    },
    {
      "epoch": 6.276439790575917,
      "grad_norm": 120.44068908691406,
      "learning_rate": 3.7280334728033472e-06,
      "loss": -155.7816,
      "step": 1500
    },
    {
      "epoch": 8.368586387434554,
      "grad_norm": 97.27143859863281,
      "learning_rate": 1.6359832635983263e-06,
      "loss": -158.0228,
      "step": 2000
    }
  ],
  "logging_steps": 500,
  "max_steps": 2390,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}